The administration of peripheral vasodilators produces hemodynamic and clinical improvement in patients with refractory congestive heart fialure. The responses to these agents in our experience, however, has varied greatly. Our previous work has established at least 3 determinant factors of importance in modulating the responses observed: (1) the dose of the vasodilator agent; (2) the degree of preservation of compensatory mechanisms which are activated by and serve to counteract the peripheral vasodilation; (3) the subvalvular left ventricular end-diastotic dimension which reflects either ventricular volume or the severity of mitral regurgitation. We propose to examine the mechanisms behind these determinant factors, how they interact to modulate the responses observed and to evaluate if other factors are operative as well. Fifty patients with severe heart failure will be evaluated by invasive hemodynamic monitoring. Each will receive 3 vasodilator drugs (hydralazine, nitroglycerin and nitroprusside) intravenously and in stepwise increments to permit construction of dose response curves. Hemodynamic variables, ventricular and regurgitant volumes, plasma catecholamines and renin activity and plasma levels of 6-keto-PGF-1 alpha will be determined prior to and during therapy; Differences in the responses observed will be analyzed in terms of variations in the determinant factors before and after therapy. Determinant factors will then be enhanced or depressed by trimethophan camsylate, phenylephrine, indomethacin or teprotide and the administration of vasodilator agents repeated; the resultant shifts in the dose response curves will be related to changes in the determinant factors. In this way, we will be able to identify specific factors (systolic and diastolic wall stress, radionuclide regurgitant fraction, neurohumoral states) which can determine the response to vasodilators in human heart failure and can be utilized clinically in selecting those patients most likely to respond favorably to vasodilator therapy.